SQL Server : outputting duplicates - sql

Question

I am using a prebuilt program so I am unable to directly edit the SQL, I can only edit snippets of it. The problem I am having is that the code is printing out the customer code for as many cases that customer has. For example, if a customer has 57 cases, it will print the customer code 57 times instead of just showing it once such as Customer Code 4 Cases 57 etc.
I was reading that you may not be able to use a Unique or Distinct function with the OVER command but I am not sure how else to make the sum work without it. Here is my code:
SET NOCOUNT ON;
DECLARE #ShowZeros nVarChar(4000);
SET #ShowZeros = 'N';
SELECT
SUM ([IC_ProductLots].[Available_Alt]) OVER(PARTITION BY [AR_Customers]. [CustomerCode]) AS [Cases]
, IC_Products.UnitOfMeasure_Alt
, SUM([IC_ProductLots].[Available_Stk]) OVER(PARTITION BY [AR_Customers]. [CustomerCode]) AS [Total Stock]
, IC_Products.UnitOfMeasure_Stk
, IC_Products.Description1
, IC_Products.ProductCode
, AR_Customers.Name
, AR_Customers.CustomerCode
FROM
((( DC_Transactions
INNER JOIN
AR_Customers ON DC_Transactions.CustomerKey = AR_Customers.CustomerKey)
INNER JOIN
IC_ProductLots ON DC_Transactions.LotKey = IC_ProductLots.LotKey)
INNER JOIN
IC_Products ON DC_Transactions.ProductKey = IC_Products.ProductKey)
WHERE
(IC_Products.ProductCode = ' 515070') AND
((CASE WHEN #ShowZeros = 'Y' or #ShowZeros = 'YES' THEN 1 ELSE
(ISNULL([IC_ProductLots].[Available_Stk],0))
END) > 0)
ORDER BY
IC_Products.ProductCode
, AR_Customers.CustomerCode
, AR_Customers.Name
Output should look similar to below:
Cases | U/M | Total Stock | Description | Cust Name | Cust Code
-----------------------------------------------------------------
57 | CS | 1779.45 | Food | Restaurant | 2
4 | CS | 120 | Dough | Bakery | 44
Currently it prints out 57 lines for customer code 2 and 4 lines for customer code 44, displaying the same information for each customer code.

Answer 1

The following seems to be more or less what you want. You say you can only partly edit your query, but well, you need the GROUP BY clause to say which records you want (i.e. one record per product and customer), so this is what mainly needs be done. Then you cannot use an OVER clause, as the aggregation group is given by GROUP BY already. I have made more changes for readability. See what edits you can do.
select
sum(l.available_alt) as cases
, p.unitofmeasure_alt
, sum(l.available_stk) as [total stock]
, p.unitofmeasure_stk
, p.description1
, p.productcode
, c.name
, c.customercode
from dc_transactions t
inner join ar_customers c on t.customerkey = c.customerkey
inner join ic_productlots l on t.lotkey = l.lotkey
inner join ic_products p on t.productkey = p.productkey
where p.productcode = ' 515070'
and (#showzeros in ('Y', 'YES') or l.available_stk > 0)
group by p.productcode, c.customercode
, p.unitofmeasure_alt, p.unitofmeasure_stk, p.description1, c.name
order by p.productcode, c.customercode;